This community is for professionals and enthusiasts of our products and services.
Share and discuss the best content and new marketing ideas, build your professional profile and become a better marketer together.
What does “expected reward over an infinite horizon” mean in a Markov Decision Process?
a) The total reward over a finite number of steps
b) The expected long-term reward considering all future decisions
c) The reward for the initial step
d) The reward for a single decision
b) The expected long-term reward considering all future decisions
In a Markov Decision Process (MDP), the expected reward over an infinite horizon refers to the total expected reward that is accumulated over an infinite number of steps, considering all future decisions made in the process. This concept helps in evaluating policies based on their long-term performance rather than focusing on a limited number of steps.